Timing- and Dose-Specific Associations of Prenatal Smoke Exposure With Newborn DNA Methylation

Page created by Martha Marshall
 
CONTINUE READING
Nicotine & Tobacco Research, 2020, 1917–1922
                                                                                              doi:10.1093/ntr/ntaa069
                                                                                                         Brief Report
                                Received October 9, 2019; Editorial Decision April 16, 2020; Accepted April 22, 2020

Brief Report

Timing- and Dose-Specific Associations of
Prenatal Smoke Exposure With Newborn DNA
Methylation

                                                                                                                                                                      Downloaded from https://academic.oup.com/ntr/article/22/10/1917/5824820 by guest on 07 December 2020
Giulietta S. Monasso MD1,2, Vincent W. V. Jaddoe MD, PhD1,2,
Johan C. de Jongste MD, PhD1,2, Liesbeth Duijts MD, PhD1,2,
Janine F. Felix MD, PhD1,2
The Generation R Study Group, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands;
1

Department of Pediatrics, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands
2

Corresponding Author: Janine F. Felix, MD, PhD, Generation R Study Group (Na-2918), Erasmus MC, University Medical Center
Rotterdam, Rotterdam, The Netherlands. Telephone: +31-10-7043405; Fax: +31-10-7044645; E-mail: j.felix@erasmusmc.nl

Abstract
Introduction: Fetal changes in DNA methylation may underlie associations of maternal smoking
during pregnancy with adverse outcomes in children. We examined critical periods and doses of
maternal smoking during pregnancy in relation to newborn DNA methylation, and associations of
paternal smoking with newborn DNA methylation.
Aims and Methods: This study was embedded in the Generation R Study, a population-based
prospective cohort study from early pregnancy onwards. We assessed parental smoking during
pregnancy using questionnaires. We analyzed associations of prenatal smoke exposure with new-
born DNA methylation at 5915 known maternal smoking-related cytosine-phosphate-guanine sites
(CpGs) in 1261 newborns using linear regression. Associations with false discovery rate-corrected
p-values < .05 were taken forward.
Results: Sustained maternal smoking was associated with newborn DNA methylation at 1391
CpGs, compared with never smoking. Neither quitting smoking early in pregnancy nor former
smoking was associated with DNA methylation, compared with never smoking. Among sustained
smokers, smoking ≥5, compared with
1918                                                                                    Nicotine & Tobacco Research, 2020, Vol. 22, No. 10

Introduction                                                              or third trimester). Among sustained smokers, we recategorized the
                                                                          reported daily cigarette dose in the third trimester into
Nicotine & Tobacco Research, 2020, Vol. 22, No. 10                                                                                                            1919

    DNA methylation levels differ per cell type; therefore, correc-                  Of these, 941 CpGs (67.6%) had lower DNA methylation levels.
tion for estimated white blood cell subtypes is needed. There is no                  Cg05575921 (AHRR) showed the strongest association, with a
consensus on which cord-blood specific reference panel should be                     6.9% lower DNA methylation level (standard error = 0.4, FDR
used for cell type estimation. To ensure robust results, independent                 p-value = 2.11E-67). Neither quitting smoking early in pregnancy
of the reference panel, we repeated the main timing analysis using                   (smallest FDR p-value = .08) nor former smoking (smallest FDR
the “Bakulski” 17 rather than the “Gervin” panel,15 as well as without               p-value = .63) was associated with newborn DNA methylation,
cell-type adjustment. The “Bakulski” panel includes the six white                    compared with never smoking in any of the models (Supplementary
blood cell subtypes from the “Gervin” panel plus nucleated red                       Table 3b and c).
blood cells.                                                                             When compared with the main model, additional adjustment
    For all analyses, we calculated the correlations between the effect              for BMI or birthweight decreased the number of associated CpGs
estimates of the main models versus all other models. For the sig-                   to 988 and 1206, respectively. For the 1391 CpGs associated with
nificant CpGs of the main model, we calculated the change in effect                  sustained maternal smoking, the mean change in effect estimates
estimates in the BMI and birthweight model.                                          was 9.5% in the BMI model and 5.1% in the birthweight model.
    We accounted for multiple testing by applying the false discovery                Results of the Bakulski-adjusted model and model without cell-type

                                                                                                                                                                       Downloaded from https://academic.oup.com/ntr/article/22/10/1917/5824820 by guest on 07 December 2020
rate (FDR) method of Benjamini and Hochberg.18 Associations with                     adjustment were very similar (Supplementary Table 3a–c). For all
FDR-corrected p-values of  .92, p < .001).
Results
Participant Characteristics                                                          Dose
Participant characteristics are shown in Table 1 and Supplementary                   Among 157 sustained smoking mothers, the third-trimester dose
Table 1. Newborns included in the analyses less often had a sus-                     ranged between 20 cigarettes/d (Supplementary Table
tained smoking mother or smoking father, compared with those not                     1). Smoking ≥5 cigarettes (N = 76) was associated with newborn
included. Furthermore, they had older parents and higher educated                    DNA methylation levels at seven CpGs in the main model (smallest
mothers with slightly lower BMI (Supplementary Table 2).                             FDR p-value = .001), when compared with smoking
1920                                                                                           Nicotine & Tobacco Research, 2020, Vol. 22, No. 10

Table 2. Associations of sustained maternal smoking during pregnancy with newborn DNA methylation, when compared with
never smoking

CpG                Mapped gene         Nearest gene (within 10 Mb)         Effect estimate (beta)      Standard error        p-value   p-value (FDR)

cg05575921          AHRR                        AHRR                              −0.069                   0.004            3.57E-71     2.11E-67
cg18316974          GFI1                        GFI1                              −0.059                   0.004            8.69E-37     2.57E-33
cg04180046          MYO1G                       MYO1G                              0.051                   0.004            1.09E-29     2.15E-26
cg12803068          MYO1G                       MYO1G                              0.068                   0.006            4.18E-28     6.18E-25
cg09935388          GFI1                        GFI1                              −0.083                   0.008            1.21E-26     1.43E-23
cg14179389          GFI1                        GFI1                              −0.064                   0.007            5.85E-22     5.77E-19
cg06338710          GFI1                        GFI1                              −0.044                   0.005            2.07E-21     1.75E-18
cg25949550          CNTNAP2                     CNTNAP2                           −0.023                   0.002            3.41E-21     2.52E-18
cg22132788          MYO1G                       MYO1G                              0.039                   0.004            1.45E-19     9.52E-17
cg12876356          GFI1                        GFI1                              −0.085                   0.010            7.68E-18     4.54E-15
cg18146737          GFI1                        GFI1                              −0.098                   0.012            5.35E-16     2.87E-13

                                                                                                                                                        Downloaded from https://academic.oup.com/ntr/article/22/10/1917/5824820 by guest on 07 December 2020
cg23067299          AHRR                        AHRR                               0.033                   0.004            1.26E-15     6.21E-13
cg19089201          MYO1G                       MYO1G                              0.028                   0.004            2.97E-15     1.35E-12
cg21161138          AHRR                        AHRR                              −0.022                   0.003            9.93E-14     4.19E-11
cg11902777          AHRR                        AHRR                              −0.015                   0.002            5.28E-13     2.08E-10
cg09662411          GFI1                        GFI1                              −0.050                   0.007            7.81E-12     2.89E-09
cg04598670          —                           AUTS2                             −0.031                   0.005            1.14E-11     3.96E-09
cg08606254          AHRR                        AHRR                               0.022                   0.003            4.62E-11     1.52E-08
cg22549041          CYP1A1                      CYP1A1                             0.055                   0.008            8.46E-11     2.63E-08
cg14817490          AHRR                        AHRR                              −0.025                   0.004            9.16E-11     2.71E-08
cg18703066          —                           —                                 −0.011                   0.002            3.15E-10     8.87E-08

Shown are all CpGs with a FDR-corrected p-value of
Nicotine & Tobacco Research, 2020, Vol. 22, No. 10                                                                                                        1921

to maternal smoking dose previously.7–9 Of these, cg05575921                    Funding
(AHRR), cg07339236 (ATP9A), cg04180046 (MYO1G), and
                                                                                The general design of the Generation R Study is made possible by financial
cg12803068 (MYO1G) showed a tendency toward dose-dependent                      support from the Erasmus MC, University Medical Centre Rotterdam,
associations in at least two studies.7–9                                        Erasmus University Rotterdam, the Netherlands Organization for Health
                                                                                Research and Development (ZonMw), the Netherlands Organization for
Paternal Smoking                                                                Scientific Research (NWO), the Ministry of Health, Welfare and Sport and
                                                                                the Ministry of Youth and Families. The EWAS data were funded by a
Like maternal smoking, paternal smoking has been associated
                                                                                grant to VWJ from the Netherlands Genomics Initiative (NGI)/Netherlands
with adverse child health outcomes.2,3,20 If these associations are
                                                                                Organization for Scientific Research (NWO) Netherlands Consortium for
mediated through DNA methylation, this could either reflect
                                                                                Healthy Aging (NCHA; project number 050-060-810), by funds from the
transmission through effects on the sperm epigenome, or in utero                Genetic Laboratory of the Department of Internal Medicine, Erasmus MC,
exposure to secondhand smoke.10 In line with a previous study,                  and by a grant from the National Institute of Child and Human Development
focusing on a small number of CpGs, we observed no associ-                      (R01HD068437). VWVJ received funding from the European Research
ations of paternal smoking with newborn DNA methylation at                      Council (ERC-2014-CoG-648916). The project was supported by funding
CpGs associated with sustained maternal smoking, independent                    from the European Union Horizon 2020 research and innovation program

                                                                                                                                                                    Downloaded from https://academic.oup.com/ntr/article/22/10/1917/5824820 by guest on 07 December 2020
of the mother’s smoking status.10 A recent full EWAS did report                 under grant agreement No 733206 (LifeCycle) and from the European Joint
evidence for associations of paternal smoking with differential                 Programming Initiative “A Healthy Diet for a Healthy Life” (JPI HDHL,
                                                                                NutriPROGRAM project, ZonMw the Netherlands no.529051022 and
DNA methylation in offspring at older ages, after adjustment
                                                                                Precise project no. P75416).
for maternal smoking.21 This indicates that there may be CpG
sites at which DNA methylation is specifically related to paternal
smoking.                                                                        Acknowledgments
                                                                                The Generation R Study is conducted by the Erasmus Medical Center in close
Strengths and Limitations                                                       collaboration with the School of Law and Faculty of Social Sciences of the
This study was embedded in a large population-based prospective                 Erasmus University Rotterdam, the Municipal Health Service Rotterdam
cohort study, and the sample size for DNA methylation was                       area, Rotterdam, the Rotterdam Homecare Foundation, Rotterdam, and
relatively large. However, this study may still not be adequately               the Stichting Trombosedienst & Artsenlaboratorium Rijnmond (STAR-
powered to find small effects. Therefore, our results should be con-            MDC), Rotterdam. We gratefully acknowledge the contribution of children
sidered as hypothesis generating. Because of power, we only in-                 and parents, general practitioners, hospitals, midwives, and pharmacies
                                                                                in Rotterdam. The study protocol was approved by the Medical Ethical
cluded CpGs previously identified in relation to sustained maternal
                                                                                Committee of the Erasmus Medical Centre, Rotterdam. Written informed con-
smoking. Consequently, we may have failed to detect CpGs spe-
                                                                                sent was obtained for all participants. The generation and management of
cifically associated with maternal smoking earlier in pregnancy or
                                                                                the Illumina 450K methylation array data (EWAS data) for the Generation
with paternal smoking. The Generation R Study provided 13% of                   R Study was executed by the Human Genotyping Facility of the Genetic
the meta-analysis sample size, so has contributed to the discovery              Laboratory of the Department of Internal Medicine, Erasmus MC, and the
of the CpGs of interest.5 As DNA methylation is tissue specific,                Netherlands. We thank Mr. Michael Verbiest, Ms. Mila Jhamai, Ms. Sarah
the relevance of findings in blood in relation to other tissues needs           Higgins, Mr. Marijn Verkerk, and Dr. Lisette Stolk for their help in creating the
further study. For example, maternal smoking during pregnancy                   EWAS database. We thank Dr. A. Teumer for his work on the quality control
has been associated with childhood asthma.1 To better understand                and normalization scripts.
the role of DNA methylation in this, analyses of DNA methyla-
tion in lung tissue, which may be more relevant for respiratory
phenotypes, should be performed. This study included relatively                 Declaration of Interests
few heavy smokers, most smokers consumed low doses and vari-                    None declared.
ation in dose was limited. Furthermore, there may be measurement
error in dose, as women may not have accurately reported small
differences in daily smoking dose. These factors may have led to                References
underestimation of any dose–response effects.                                   1. Cnattingius S. The epidemiology of smoking during pregnancy: smoking
                                                                                   prevalence, maternal characteristics, and pregnancy outcomes. Nicotine
                                                                                   Tob Res. 2004;6(suppl 2):S125–S140.
Conclusion                                                                      2. Abraham M, Alramadhan S, Iniguez C, et al. A systematic review of ma-
Our results indicate that CpGs robustly associated with sustained                  ternal smoking during pregnancy and fetal measurements with meta-
maternal smoking are not associated with maternal smoking earlier                  analysis. PLoS One. 2017;12(2):e0170946.
                                                                                3. Jaddoe VWV, Troe EWM, Hofman A, et al. Active and passive ma-
in pregnancy or with paternal smoking and that the association
                                                                                   ternal smoking during pregnancy and the risks of low birthweight and
with sustained maternal smoking shows dose effects at some CpGs.
                                                                                   preterm birth: the Generation R Study. Paediatr Perinat Epidemiol.
Further research is needed to study the relevance of DNA methyla-
                                                                                   2008;22(2):162–171.
tion for clinical outcomes.                                                     4. McCowan LM, Dekker GA, Chan E, et al.; SCOPE consortium.
                                                                                   Spontaneous preterm birth and small for gestational age infants in women
                                                                                   who stop smoking early in pregnancy: prospective cohort study. BMJ.
Supplementary Material                                                             2009;338:b1081.
A Contributorship Form detailing each author’s specific involvement with this   5. Joubert BR, Felix JF, Yousefi P, et al. DNA Methylation in newborns and
content, as well as any supplementary data, are available online at https://       maternal smoking in pregnancy: genome-wide consortium meta-analysis.
academic.oup.com/ntr.                                                              Am J Hum Genet. 2016;98(4):680–696.
1922                                                                                            Nicotine & Tobacco Research, 2020, Vol. 22, No. 10

6. Miyake K, Kawaguchi A, Miura R, et al. Association between DNA                14. Niklasson A, Ericson A, Fryer JG, Karlberg J, Lawrence C, Karlberg P. An
    methylation in cord blood and maternal smoking: the Hokkaido Study on            update of the Swedish reference standards for weight, length and head cir-
    Environment and Children’s Health. Sci Rep. 2018;8(1):5654.                      cumference at birth for given gestational age (1977–1981). Acta Paediatr
7. Richmond RC, Simpkin AJ, Woodward G, et al. Prenatal exposure to ma-              Scand. 1991;80(8–9):756–762.
    ternal smoking and offspring DNA methylation across the lifecourse: find-    15. Gervin K, Page CM, Aass HC, et al. Cell type specific DNA methylation in
    ings from the Avon Longitudinal Study of Parents and Children (ALSPAC).          cord blood: a 450K-reference data set and cell count-based validation of
    Hum Mol Genet. 2015;24(8):2201–2217.                                             estimated cell type composition. Epigenetics. 2016;11(9):690–698.
8. Joubert BR, Håberg SE, Nilsen RM, et al. 450K epigenome-wide scan identi-     16. Houseman EA, Molitor J, Marsit CJ. Reference-free cell mixture ad-
    fies differential DNA methylation in newborns related to maternal smoking        justments in analysis of DNA methylation data. Bioinformatics.
    during pregnancy. Environ Health Perspect. 2012;120(10):1425–1431.               2014;30(10):1431–1439.
9. Markunas CA, Xu Z, Harlid S, et al. Identification of DNA methylation         17. Bakulski KM FJ, Andrews SV, Yang J, Brown S, McKenney LS et al. DNA
    changes in newborns related to maternal smoking during pregnancy.                methylation of cord blood cell types: applications for mixed cell birth
    Environ Health Perspect. 2014;122(10):1147–1153.                                 studies. Epigenetics. 2016;11(5):354–362.
10. Joubert BR, Håberg SE, Bell DA, et al. Maternal smoking and DNA              18. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a prac-
    methylation in newborns: in utero effect or epigenetic inheritance? Cancer       tical and powerful approach to multiple testing. J.R. Statist. Soc. B.

                                                                                                                                                                  Downloaded from https://academic.oup.com/ntr/article/22/10/1917/5824820 by guest on 07 December 2020
    Epidemiol Biomarkers Prev. 2014;23(6):1007–1017.                                 1995;57(1):289–300.
11. Kooijman MN, Kruithof CJ, Van Duijn CM, et al. The Generation R Study:       19. Joehanes R, Just AC, Marioni RE, et al. Epigenetic signatures of cigarette
    design and cohort update 2017. Eur J Epidemiol. 2016;31(12):1243–1264.           smoking. Circ Cardiovasc Genet. 2016;9(5):436–447.
12. Durmuş B, Heppe DHM, Taal HR, et al. Parental smoking during preg-           20. Windham GC, Hopkins B, Fenster L, Swan SH. Prenatal active or passive
    nancy and total and abdominal fat distribution in school-age children: the       tobacco smoke exposure and the risk of preterm delivery or low birth
    Generation R Study. Int J Obes (Lond). 2014;38:966.                              weight. Epidemiology. 2000;11(4):427–433.
13. Lehne B, Drong AW, Loh M, et al. A coherent approach for analysis of the     21. Mørkve Knudsen GT, Rezwan FI, Johannessen A, et al. Epigenome-
    Illumina HumanMethylation450 BeadChip improves data quality and per-             wide association of father’s smoking with offspring DNA methylation: a
    formance in epigenome-wide association studies. Genome Biol. 2015;16:37.         hypothesis-generating study. Environ Epigenet. 2019;5(4):dvz023.
You can also read